← Back to community benchmarks

Gemma4-26B-MoE-Q8

M5 Max (40c) · 48 GB · 8bit · 2026-07-21
Performance
16k
tokens
3,417
PP tok/s
67.5
TG tok/s
4794
TTFT (ms)
27.8
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.5.2rc1
macOS macOS 26.5.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
4k 3,595 75.0 26.9 GB view
16k 3,417 67.5 27.8 GB current